The proto-oncogene HER2/neu (C-erbB-2) encodes a transmembrane tyrosine kinase growth factor receptor. The name for the HER2 protein is derived from “Human Epidermal growth factor Receptor,” as it features substantial homology with the epidermal growth factor receptor (EGFR). Overexpression of this transmembrane receptor has been found in ˜30% of breast cancers (Slamon et al., 1987), amounting to as many as 60-thousand cases a year in the United States. To date, a wide variety of clinical studies, including more than 15,000 patients, have evaluated the relationship between HER2/neu abnormalities and breast cancer outcome, including more than 47 publications concerning the association of gene and/or protein abnormalities with prognosis (Ross et al., 1999). These studies clearly indicate that HER2 protein overexpression is associated with an adverse outcome in breast cancer.
Overexpression of the oncogene Her2 (neu/ErbB2) has been found in ˜30% of breast tumors (Slamon et al., 1987; Ross et al., 1999). It is unclear why overexpression of this transmembrane receptor occurs in breast cancers, but patients who have Her2 excesses typically have more aggressive disease with enhanced metastasis and increased resistance to chemotherapy. Monoclonal antibodies against the Her2 protein have been successful in treating these Her2-positive patients. A humanized antibody called Herceptin® has demonstrated tumor inhibitory and chemosensitizing effects in clinical studies and is the only drug that FDA has approved for treatment of Her2-overexpressing breast tumors (Drebin et al.,1986; Drebin et al., 1985). The clinical success of the Her2-antibody therapy was an excellent example of the translation of basic cancer biology into clinical cancer treatment. However, the antibody therapy alone may not be ideal for therapeutic intervention of Her2-overexpressing breast cancers. In theory, downregulation of Her2 may be accomplished efficiently by inhibiting the expression of the Her2 gene rather than targeting elevated levels of the Her2 proteins that are already overexpressed. By analogy with treatments for AIDS, cocktails of drugs—each with a different mechanisms of action—might also be more effective at achieving complete remission of breast tumors. Thus, discovering a means to provide external control over Her2 expression, particularly through small organic molecules, remains appealing.
The overexpression of Her2 can be accomplished through a combination of two different mechanisms: gene duplication and increased transcription. In both cases, excess Her2 mRNA is produced, and this process can be controlled at the level of gene transcription. In search of transcription factors that activate the Her2 gene in breast cancers, Chang et al. and others have identified the Ets factor ESX (epithelial-restricted with serine box) (Chang et al., 1997 ). ESX is overexpressed in Her2-overexpressing breast cancer cells, and it binds and strongly transactivates the promoter of the Her2 gene. Deletion of the ESX-binding element from the Her2 promoter completely abolishes its activation, suggesting the importance of ESX in activating the Her2 gene in cells (Scott et al., 1994). The expression of ESX is epithelial specific and high in mammalian gland, trachea, prostate, and stomach (Andreoli et al., 1997). ESX may be a unique transcription factor that controls expression of Her2 in these tissues.
Current drug therapy addresses only about 500 molecular targets. Cell membrane receptors and enzymes account for ˜80% of all current drug targets, and there are few drug targets in the nucleus except nuclear receptors (and DNA in the case of cancer chemotherapy) (Drews et al., 2000). Discovering new molecular targets in the nucleus would extend the scope of drug targets and might provide alternative therapeutic strategies to treat major human diseases. For instance, recent discovery of histone deacetylases as a potential target for cancer therapy had tremendous impacts on the drug discovery research (Kwon et al., 1998; Hassing et al., 1997; Taunton et al., 1996; Lin et al., 1998). The potentiality of transcriptional co-activators as a drug target has never been explored due to the lack of expected therapeutic phenotypes. A subset of co-activators, although not all, may serve as a target for pharmaceutical intervention.
Complete analysis of the human genome is anticipated to produce an unprecedented number of potential drug targets. Among these genomic pseudo-targets, the “relatively easy” targets such as GPCRs or enzymes will certainly be an immediate focus in pharmaceutical industries. However, more challenging genomic targets including protein-protein interactions need to be assessed for a leap of pharmaceutical development in the future.
Protein-protein interactions are harder to target by small organic molecules than enzymes or nuclear hormone receptors. Protein-protein binding typically occurs over a relatively large surface area, and the binding surfaces between two proteins tend to be flat and often lack in pockets that might provide binding sites suited for small organic molecules. Nevertheless, protein-protein interfaces vary widely in nature from one to another, and some are likely to present better druggability than others. A good example is the interaction between the somatostatin receptor and β-turn peptide ligands.
Recent studies suggest that the protein-protein interactions that are mediated by short α-helical segments of proteins are similarly tractable to inhibition by small nonpeptidic molecules. Helical peptide segments of proteins are responsible for a number of biologically important protein associations in the fields of signal transduction and gene transcription. The interaction between the two cancer-linked nuclear proteins, ESX (an epithelial-specific transcription factor) and Sur-2/DRIP130 (a Ras-linked subunit of the human mediator complex) is required for the overexpression of the Her2 oncogene in malignant breast cancer cells and thus serves as a potential therapeutic target for Her2-positive breast cancers amounting to 60,000 cases per year in the United States. The interaction is mediated by one face of an eight-amino acid-helical region in the transcriptional activation domain of ESX (SEQ ID NO: 1 Ser-Trp-Ile-Ile-Glu-Leu-Leu-Glu), and the tryptophan residue in the hydrophobic face of the helix makes a unique contribution to the specificity of the interaction. The relatively small size of the interface and the importance of the tryptophan residue suggested the existence of small-molecule inhibitors in a chemical library enriched in the structural families of indole, benzimidazole, and benzodiazepin—indole-mimicking π-electron-rich pharmacores found in bioavailable drugs.